AIX

Contributors: Yadagiri Rajaboina, Vinayak karne.

In AIX when Fibre Channel(FC) SCSI device driver detects a FC link failure event, driver waits approximately for 15 seconds before flushing(switching) the I/Os to alternative MPIO paths. For example, if the driver receives a Registered State Change Notification (RSCN) from the switch, this could indicate a link loss between a remote storage port and switch. The driver queries to see if the storage port has joined the fabric for 15 seconds. If the storage port does not join the fabric, then driver flushes the I/O’s to upper layers. This causes delay in switching I/Os to alternate MPIO paths.

Note : X in above figure indicates FC link failure.

To quickly flush the I/Os starting from 4 seconds, for host or storage side link failure scenarios, the AIX FC SCSI driver stack is enhanced from AIX 7.3 TL1 release onward.

This functionality is controlled through a new ODM attribute “fast_lnk_recov” at protocol driver level for FC HBAs that has data rates equal to or more than 16Gbps. This attribute is concurrently updatable i.e the attribute value can be changed while the device is configured. The supported values of this attribute are “yes”,” no” with default value is set to “no”.

A sample output from AIX where this ODM attribute is updated concurrently is explained below.

This tunable parameter is only used for a multi path I/O (MPIO) environment where another path to fail over is available. As a best practice, have an alternative path through another Fibre Channel host bus adapter (HBA) to handle HBA hardware failure scenarios. In a non-MPIO environment, it is advisory that the default settings are left unchanged.

End of the Document